Piscataway, NJ — Arthur Augustyn reports in NJBIZ that as companies get larger, so do their challenges. For big data problems, the difficulty isn’t finding a solution but understanding the scale of an issue.
Computers have been an integral part of simplifying company problems and finding solutions at faster rates than individuals. But as the problems of businesses get large and more complex, computers also need to beef up.
Rutgers University’s Office of Advanced Research Computing rolled out its most advanced technology yet in July. “Amarel,” is a computing system meant to assist the research community with tackling problems that require a massive amount of calculations and variables.
Computers have been an integral part of simplifying company problems and finding solutions at faster rates than individuals. But as the problems of businesses get large and more complex, computers also need to beef up.
Rutgers University’s Office of Advanced Research Computing rolled out its most advanced technology yet in July. “Amarel,” is a computing system meant to assist the research community with tackling problems that require a massive amount of calculations and variables.
Officially, Amarel is a “high performance community cluster,” named after a founder of Rutgers’ Computer Science Department Dr. Saul Amarel. But it wouldn’t be inaccurate to call Amarel a “supercomputer.”
Amarel’s processing power allows the system to tackle problems that ordinary computers aren’t equipped to handle.
For example, big data analytics are useful to businesses that want to study customer trends but compiling that information can take days or weeks, and applying it to a theoretical model can take even longer.
“The models are becoming much more sophisticated,” said James Barr von Oehsen, associate vice president for Office of Advanced Research Computing. The data sets are getting much bigger, he said, and involve types of analysis that couldn’t be done on a desktop.
Big data and data analytics have become popular phrases in recent years as companies take more interest in understanding consumer habits on a larger scale.
One field that’s taking advantage of Amarel’s technology is genetics. Data for one set of human DNA is a massive amount of information.
“We have molecular tools that let us look at that DNA and figure out the sequence, but in order to ask downstream questions we have to take that DNA and figure out where it goes into the genome as a whole,” said Elizabeth Snyder, assistant professor of animal science at Rutgers-New Brunswick. “It’d be completely impossible without [Amarel].”
According to von Oehsen, one genetics researcher used Amarel to work a computation that typically took 12 hours on a regular desktop. Amarel churned it out in 6 seconds.
In the medical world, high resolution MRIs are used to find cancer in patients. Typically, radiologists look at an MRI scan and make a diagnosis, but with advanced computing, a machine like Amarel could look at a high resolution image and examine potential points of concern pixel by pixel.
Amarel is capable of “machine learning,” a process that teaches the system what to look for. In an MRI scan, the machine can learn from repeat experience what constitutes an image that could indicate a tumor.
Although the learning process is a new development for advanced computing systems, it’s far away from the concept of artificial intelligence that’s been popularized by mainstream media.
Amarel’s uses are not strictly for the Science Technology Engineering Math (STEM) fields.
As the world gets digitized, the uses of a computing system expand. For example, a historian could tag a specific era to research and pull all written sources from that timeframe.
Instead of relying on the available books at a local library, they can utilize the entire works of an online repository.
For the full NJBIZ story, click here.
Follow Augustyn on Twitter @ArthurAugustyn.
Amarel’s processing power allows the system to tackle problems that ordinary computers aren’t equipped to handle.
For example, big data analytics are useful to businesses that want to study customer trends but compiling that information can take days or weeks, and applying it to a theoretical model can take even longer.
“The models are becoming much more sophisticated,” said James Barr von Oehsen, associate vice president for Office of Advanced Research Computing. The data sets are getting much bigger, he said, and involve types of analysis that couldn’t be done on a desktop.
Big data and data analytics have become popular phrases in recent years as companies take more interest in understanding consumer habits on a larger scale.
One field that’s taking advantage of Amarel’s technology is genetics. Data for one set of human DNA is a massive amount of information.
“We have molecular tools that let us look at that DNA and figure out the sequence, but in order to ask downstream questions we have to take that DNA and figure out where it goes into the genome as a whole,” said Elizabeth Snyder, assistant professor of animal science at Rutgers-New Brunswick. “It’d be completely impossible without [Amarel].”
According to von Oehsen, one genetics researcher used Amarel to work a computation that typically took 12 hours on a regular desktop. Amarel churned it out in 6 seconds.
In the medical world, high resolution MRIs are used to find cancer in patients. Typically, radiologists look at an MRI scan and make a diagnosis, but with advanced computing, a machine like Amarel could look at a high resolution image and examine potential points of concern pixel by pixel.
Amarel is capable of “machine learning,” a process that teaches the system what to look for. In an MRI scan, the machine can learn from repeat experience what constitutes an image that could indicate a tumor.
Although the learning process is a new development for advanced computing systems, it’s far away from the concept of artificial intelligence that’s been popularized by mainstream media.
Amarel’s uses are not strictly for the Science Technology Engineering Math (STEM) fields.
As the world gets digitized, the uses of a computing system expand. For example, a historian could tag a specific era to research and pull all written sources from that timeframe.
Instead of relying on the available books at a local library, they can utilize the entire works of an online repository.
For the full NJBIZ story, click here.
Follow Augustyn on Twitter @ArthurAugustyn.
